Abstract: On the basis of Laplaian score (LS) method, a supervised laplaian score (SLS) feature selection method is proposed. This method takes into account the data label information and local geometric structure, thus the problem of setting the neighbor graph parameters in LS method is avoided. Combined SLS with principal component analysis (PCA), a fault diagnosis method of rolling bearings is put forward. The feature of vibration signals of a rolling bearing is extracted in time domain and frequency domain, from which an initial feature vector is formed. By using SLS method to select features, fault feature vectors are obtained. Then, the PCA method is used to reduce the dimension of fault feature vectors and the K-nearest neighbor (KNN) method is used as a fault feature classifier to recognize different fault types of a rolling bearing. Application examples show that this method can be used to extract the features of vibration signals of rolling bearings and diagnosis the fault of rolling bearings effectively.

Key words: feature selection;supervised Laplaian score;principal component analysis;fault diagnosis